Spotify will use Gemini to create playlists based on your surroundings

The convergence of artificial intelligence and wearable devices is taking very rapid steps, and the main players in the tech sector are already moving to avoid being caught unprepared. With the imminent launch of smart glasses based on the platform Android XR, the market forecasts of Google point to the sale of around 2 million units just for this year.

In this context, Spotify is taking action well in advance to integrate its services within this unprecedented ecosystem. The Swedish company is working to leverage Mountain View’s infrastructure, firmly focusing on the advanced capabilities of the artificial intelligence model Gemini.

Spotify and Gemini, a soundtrack generated by the environment

The most interesting aspect to emerge from the ongoing development concerns the implementation of multimodal functionalities of artificial intelligence, designed to adapt music playback to the surrounding environment.

Analyzing the source code of Spotify version 9.1.66.1259, strings of text have been discovered indicating a direct interaction between the cameras integrated into the glasses and the streaming platform.

Users will have the possibility to utter a specific voice command, such as “Hey Gemini, create a soundtrack for what I’m looking at“. At that point, the system will analyze the surrounding environment and select context-appropriate tracks.

An enclosed environment like a gym could automatically trigger an energetic music selection, while observing a natural landscape or a sunset would suggest much calmer and more relaxing melodies.

Voice interaction is the future?

Beyond visual space analysis, linking your Google account will allow you to manage your entire music library without ever using your hands or pulling the smartphone from your pocket.

The discovered code portions indicate that the application will provide a sort of built-in tutorial, explaining to users how to maximize the use of the software through voice commands.

Direct requests such as “Hey Gemini, create a playlist with my favorite artists” will enable instant personalized selections, relying entirely on the virtual assistant’s processing capabilities for real-time track search.

Considering the physical constraints and the absence of traditional screens for manual input on the Android XR platform, the use of an advanced assistant proves to be a nearly mandatory technical step to ensure acceptable usability.

Third-party software will inevitably have to rely on conversational capabilities to enable navigation between the various functions.

Field verification remains to assess the responsiveness and real accuracy of this environmental analysis function when the hardware reaches store shelves, but the premises indicate a very clear direction toward increasingly personalized listening, closely tied to everyday situations.